Chasing Elusive Memory Bugs in GPU Programs

• URL: http://arxiv.org/abs/2601.21552v1
• Date: Thu, 29 Jan 2026 11:13:36 GMT
• Title: Chasing Elusive Memory Bugs in GPU Programs
• Abstract summary: Memory safety bugs, such as out-of-bound accesses (OOB) in GPU programs, can compromise the security and reliability of GPU-accelerated software.<n>We build SCuBA, a first-of-its-kind compile-time technique that analyzes CPU and GPU code to capture semantic relations (if present)<n>Compared to NVIDIA's Compute Sanitizer that misses 45 elusive memory bugs across 20 programs, SCuBA misses none with no false alarms.
• Score: 2.7752002595974172
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Memory safety bugs, such as out-of-bound accesses (OOB) in GPU programs, can compromise the security and reliability of GPU-accelerated software. We report the existence of input-dependent OOBs in the wild that manifest only under specific inputs. All existing tools to detect OOBs in GPU programs rely on runtime techniques that require an OOB to manifest for detection. Thus, input-dependent OOBs elude them. We also discover intra-allocation OOBs that arise in the presence of logical partitioning of a memory allocation into multiple data structures. Existing techniques are oblivious to the possibility of such OOBs. We make a key observation that the presence (or absence) of semantic relations among program variables, which determines the size of allocations (CPU code) and those calculating offsets into memory allocations (GPU code), helps identify the absence (or presence) of OOBs. We build SCuBA, a first-of-its-kind compile-time technique that analyzes CPU and GPU code to capture such semantic relations (if present). It uses a SAT solver to check if an OOB access is possible under any input, given the captured relations expressed as constraints. It further analyzes GPU code to track logical partitioning of memory allocations for detecting intra-allocation OOB. Compared to NVIDIA's Compute Sanitizer that misses 45 elusive memory bugs across 20 programs, SCuBA misses none with no false alarms.

Related papers

• Towards Standardized Evaluation of GPU Memory Safety with GMSBench [9.227375465245993]
  We present GMSBench, a GPU memory safety benchmark designed to evaluate a broad range of memory safety violations.<n>GMSBench comprises 149 self-contained tests spanning spatial, temporal, and errors.<n>We demonstrate the utility of GMSBench by evaluating Compute Sanitizer, a widely used GPU memory error detection tool.
  arXiv  Detail & Related papers  (2026-09-08T15:11:57Z)
• Joint Agent Memory and Exploration Learning via Novelty Signals [46.41306111727712]
  In open-ended environments, exploration is fundamental for autonomous agents.<n>We introduce textbfJoint textbfAgent textbfMemory and textbfExploration textbfLearning (textbfJAMEL)<n>textbfJAMEL is a framework that trains agentic memory and exploration policy together through novelty-driven interaction.
  arXiv  Detail & Related papers  (2026-06-01T01:20:26Z)
• Understand and Accelerate Memory Processing Pipeline for Disaggregated LLM Inference [45.78442013449325]
  Large language models rely on efficient long-context processing and generation mechanisms to support complex reasoning.<n>We show that these optimizations can be unified into a four-step memory processing pipeline: Prepare Memory, Compute Relevancy, Retrieval, and Apply to Inference.<n>Motivated by this insight, we argue that textbfheterogeneous systems are well-suited to accelerate memory processing and thus end-to-end inference.
  arXiv  Detail & Related papers  (2026-03-30T21:03:39Z)
• Hunting CUDA Bugs at Scale with cuFuzz [2.0745580313629097]
  GPU programs are prone to memory-safety and bugs that evade static analysis.<n>While fuzz-testing, combined with dynamic error checking tools, offers a plausible solution, it remains underutilized for GPUs.<n>We present cuFuzz, the first instrument and first NV-oriented fuzzer that makes fuzzing practical.
  arXiv  Detail & Related papers  (2026-03-12T22:06:11Z)
• GPU-Fuzz: Finding Memory Errors in Deep Learning Frameworks [16.88852734850369]
  GPU memory errors are a critical threat to deep learning (DL) frameworks, leading to crashes or even security issues.<n>We introduce GPU-Fuzz, a fuzzer locating these issues efficiently by modeling operator parameters as formal constraints.
  arXiv  Detail & Related papers  (2026-02-11T03:32:43Z)
• How Would Oblivious Memory Boost Graph Analytics on Trusted Processors? [8.661898399197089]
  We focus on graph analytics, an important application vulnerable to access-pattern attacks.<n>With a co-design between storage structure and algorithms, our prototype system is 100x faster than baselines given an OM sized around the per-core cache.<n>This gives insights into equipping trusted processors with OM.
  arXiv  Detail & Related papers  (2025-12-30T14:28:29Z)
• Valori: A Deterministic Memory Substrate for AI Systems [0.0]
  Valori is a deterministic AI memory substrate that replaces floating-point memory operations with fixed-point arithmetic.<n>We show how Valori enforces determinism at the memory boundary.<n>Our results suggest that deterministic memory is a necessary primitive for trustworthy AI systems.
  arXiv  Detail & Related papers  (2025-12-25T06:04:04Z)
• xMem: A CPU-Based Approach for Accurate Estimation of GPU Memory in Deep Learning Training Workloads [2.2991119948183525]
  estimation of how much GPU memory a job will require is fundamental to enabling advanced scheduling and GPU sharing.<n>We propose xMem, a novel framework that leverages CPU-only dynamic analysis to accurately estimate peak GPU memory requirements.<n>The analysis of 5209 runs, which includes ANOVA and Monte Carlo results, highlights xMem's benefits.
  arXiv  Detail & Related papers  (2025-10-23T23:16:27Z)
• HeadInfer: Memory-Efficient LLM Inference by Head-wise Offloading [79.38548165722229]
  HEADINFER offloads the KV cache to CPU RAM while avoiding the need to fully store the KV cache for any transformer layer on the GPU.<n>We demonstrate HEADINFER maintains computational efficiency while significantly reducing memory footprint.
  arXiv  Detail & Related papers  (2025-02-18T06:26:05Z)
• Memory-Efficient Training for Deep Speaker Embedding Learning in Speaker Verification [50.596077598766975]
  We explore a memory-efficient training strategy for deep speaker embedding learning in resource-constrained scenarios.<n>For activations, we design two types of reversible neural networks which eliminate the need to store intermediate activations.<n>For states, we introduce a dynamic quantization approach that replaces the original 32-bit floating-point values with a dynamic tree-based 8-bit data type.
  arXiv  Detail & Related papers  (2024-12-02T06:57:46Z)
• ThunderKittens: Simple, Fast, and Adorable AI Kernels [43.32681787348603]
  We present ThunderKittens (TK), a framework for writing performant AI kernels while remaining easy to use and maintain. We show the value of TK by providing kernels that match or outperform prior kernels for a range of AI operations.
  arXiv  Detail & Related papers  (2024-10-27T10:07:16Z)
• KGym: A Platform and Dataset to Benchmark Large Language Models on Linux Kernel Crash Resolution [59.20933707301566]
  Large Language Models (LLMs) are consistently improving at increasingly realistic software engineering (SE) tasks. In real-world software stacks, significant SE effort is spent developing foundational system software like the Linux kernel. To evaluate if ML models are useful while developing such large-scale systems-level software, we introduce kGym and kBench.
  arXiv  Detail & Related papers  (2024-07-02T21:44:22Z)
• Host-Based Allocators for Device Memory [1.2289361708127877]
  We pose a model where the allocation algorithm runs on host memory but allocates device memory and so incur the following constraint: the allocator can't read the memory it is allocating. This means we are unable to use boundary tags, which is a concept that has been ubiquitous in nearly every allocation algorithm. In this paper, we propose alternate algorithms to work around this constraint, and discuss in general the implications of this system model.
  arXiv  Detail & Related papers  (2024-05-11T19:28:37Z)
• Recurrent Dynamic Embedding for Video Object Segmentation [54.52527157232795]
  We propose a Recurrent Dynamic Embedding (RDE) to build a memory bank of constant size. We propose an unbiased guidance loss during the training stage, which makes SAM more robust in long videos. We also design a novel self-correction strategy so that the network can repair the embeddings of masks with different qualities in the memory bank.
  arXiv  Detail & Related papers  (2022-05-08T02:24:43Z)
• Semiring Primitives for Sparse Neighborhood Methods on the GPU [16.56995698312561]
  We show that a sparse semiring primitive can be flexible enough to support a wide range of critical distance measures. This primitive is a foundational component for enabling many neighborhood-based information retrieval and machine learning algorithms to accept sparse input.
  arXiv  Detail & Related papers  (2021-04-13T17:05:03Z)
• DMV: Visual Object Tracking via Part-level Dense Memory and Voting-based Retrieval [61.366644088881735]
  We propose a novel memory-based tracker via part-level dense memory and voting-based retrieval, called DMV. We also propose a novel voting mechanism for the memory reading to filter out unreliable information in the memory.
  arXiv  Detail & Related papers  (2020-03-20T10:05:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.